1

Topic: Done all calibrations, still no perfect circles

Hi guys,
after some months of following Soliforum topics, hints, advices etc... I'm still not able to achieve perfect circles.

I finally manage to design by myself a calibration circle for 1 Euro coin in Creo Elements. The inner circle is 23.25mm, as the Euro coin, the outer circle is 43.25mm. Printed at 195C, 0.2963mm layer height, three perimeters, 20% honeycomb infill, 3 bottom solid layers, 4 top solid layers.
Speed 60mm/s for perimeters, 35mm/s for small perimeters, 45mm/s for external perimeters, 80mm/s for infill, 60mm/s for solid infill, 45mm/s for top solid infill, 70mm/s for non speed moves, first layer speed 30%

As you can see from the picture below, the Euro coin fits nicely with a bit of push, but not so much. However you can see the circle is not rounded in the top left diagonal and bottom right diagonal. The infill exhibits the same issue but on the opposite sides.

My belts have been evenly tightened using a 100g lock hung with a wire in the middle of the belt length. I also progressively tightened the belts until I didn't saw any belt movement near the pullyes when changing direction.
I also tried to increase small belt tension until I saw the Y-axis rod bending slightly.

Printer is not making any weird noise, carriage rods has been lubricated with lithium grease, I also change back pulleys position to ensure the belts are running parallel to the frame and the back rod stays in place when rotating.

Any other hints? I understand the X axis belt has no tension adjustment, right?

http://www.nicosoft.it/pictures/calibration-circle.jpg

2

Re: Done all calibrations, still no perfect circles

That looks pretty damn good really, about as close as I was able to get things using my Australian 10c coin test smile

I'm not sure just how close you can be expected to get this style of printer. Perhaps with lead screws like pro CNC machines use rather than pulleys the tolerances would be close enough.

3

Re: Done all calibrations, still no perfect circles

Loosen the long belts and see if that makes a difference.  When measuring Y backlash with a dial indicator I found that tightening the belts actually increased backlash probably because of increased friction on the idler pulleys and bushings from the tension.

Even when I got backlash to almost 0 according to the indicator I still didn't have perfect circles.  It could be that is the best you can get without upgrading the belts  and pulleys.

4

Re: Done all calibrations, still no perfect circles

agentsmith wrote:

Hi guys,
after some months of following Soliforum topics, hints, advices etc... I'm still not able to achieve perfect circles.

I finally manage to design by myself a calibration circle for 1 Euro coin in Creo Elements. The inner circle is 23.25mm, as the Euro coin, the outer circle is 43.25mm. Printed at 195C, 0.2963mm layer height, three perimeters, 20% honeycomb infill, 3 bottom solid layers, 4 top solid layers.
Speed 60mm/s for perimeters, 35mm/s for small perimeters, 45mm/s for external perimeters, 80mm/s for infill, 60mm/s for solid infill, 45mm/s for top solid infill, 70mm/s for non speed moves, first layer speed 30%

As you can see from the picture below, the Euro coin fits nicely with a bit of push, but not so much. However you can see the circle is not rounded in the top left diagonal and bottom right diagonal. The infill exhibits the same issue but on the opposite sides.

My belts have been evenly tightened using a 100g lock hung with a wire in the middle of the belt length. I also progressively tightened the belts until I didn't saw any belt movement near the pullyes when changing direction.
I also tried to increase small belt tension until I saw the Y-axis rod bending slightly.

Printer is not making any weird noise, carriage rods has been lubricated with lithium grease, I also change back pulleys position to ensure the belts are running parallel to the frame and the back rod stays in place when rotating.

Any other hints? I understand the X axis belt has no tension adjustment, right?

http://www.nicosoft.it/pictures/calibration-circle.jpg

Doesn't look so bad.

Former Solidoodle employee, no longer associated with the company.

5 (edited by Gomisan 2013-09-06 15:06:48)

Re: Done all calibrations, still no perfect circles

one of mine for comparison...

Post's attachments

2013-08-22 18.34.09.jpg
2013-08-22 18.34.09.jpg 2.08 mb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

6

Re: Done all calibrations, still no perfect circles

Thank you everybody for the feedback. I will try playing with the tension as Ian suggested.

While I can accept this level of "roundness", I'd like to have a full infill like Gomisan circle. You see that mine has some holes near the inner circle... Any tricks? smile

7

Re: Done all calibrations, still no perfect circles

Don't forget to check the tension/backlash on the y axis motor belt (small belt).  I was having a similar problem as you, slowed my printing speed and saw that the small belt was not tensioned enough. Tightened it up and my circles are about perfect.

8

Re: Done all calibrations, still no perfect circles

Agentsmith: The x axis is adjustable for tension just not as noticeable, just look on the left lower side of the print head...

9

Re: Done all calibrations, still no perfect circles

To Shotline: thank you for the advice, yes I already increased the small belt tension, I can do more but then the back rod would start bending. Is yours bent? Are there any issues in keeping this rod bent for long time?

To Ronsii: thank you for the tip! Is it enough to screw a bit more the hex screw I see on the left side of the print head adjacent to the X axis belt?

10 (edited by pirvan 2013-09-08 17:08:33)

Re: Done all calibrations, still no perfect circles

I too am having some problems printing nicely formed holes.  The problem gets worse as the size of the hole gets smaller.  Bigger holes are more accurate (or maybe the problem is not as obvious on the larger size). 

I tried the Solidoodle method of calibration, I also tried tweaking the belt tension, I even tried chaging things in software by fooling around with the stepping values, but this is about as good as it gets for me.

Take a look at the image below, the larger holes are OK, but the smaller ones are get worse.  The stretch occurs diagonally (upper left to lower right)

http://www.soliforum.com/misc.php?action=pun_attachment&item=3339

Edit:  Here is a picture that shows exactly where the gaps are.  BTW, the US quarter is about the same size as the Euro in the OP.

http://www.soliforum.com/misc.php?action=pun_attachment&item=3341

Post's attachments

holes 2.jpg
holes 2.jpg 146.79 kb, file has never been downloaded. 

holes copy.jpg
holes copy.jpg 332.89 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.
To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

11

Re: Done all calibrations, still no perfect circles

I also have almost imperceptible slits( .002-.005") max dimension. They also appear only at the 11 and 5 oclock positions.
This is curious and most likely some imbalance created by the weight of the y carriage being different from one side to the other.
The location suggests that when the x axis changes direction from front to back and from back to front the imbalance effects the location of the nozzle. That's what I am considering now. I wonder if I add weight to the left side of the y carriage would it make a difference. The other possibility is that the single bearing on the left side allows some wobble while the double bearing on the right side holds it in correct orientation better. I wonder if a redesign of the left carriage bearing to make it a double bearing spread the same distance would help.   ..  Your thoughts.

12

Re: Done all calibrations, still no perfect circles

Hi supraflyer, you might also check the 'straightness' of both x and y rods by rotating them(one at a time) 90 or 180 degrees in place and see if that changes direction or placement of you gaps.

13

Re: Done all calibrations, still no perfect circles

Seems a common issue , the distortion of my circles is exactly the same, which could indicate a problem inherent in the design. I like the idea of rotating the rods as a troubleshooting step, but for my needs it more hassle than its worth.

14

Re: Done all calibrations, still no perfect circles

agentsmith wrote:

To Shotline: thank you for the advice, yes I already increased the small belt tension, I can do more but then the back rod would start bending. Is yours bent? Are there any issues in keeping this rod bent for long time?

To Ronsii: thank you for the tip! Is it enough to screw a bit more the hex screw I see on the left side of the print head adjacent to the X axis belt?

No my rod isn't bent and if you are bending the rod I would think you have to much tension.

Another thing I did (may have helped, maybe not) is that you know how the belt tends to ride at the side of the toothed pulley?  I moved the pulley over a bit so that when you run the y axis back and forth the belt doesn't touch the sides of the pulley.  Move the pulley over a bit and slide the belt to the middle of it and move the axis back and forth, keep adjusting until it doesn't touch the sides of the pulley.

Another thing I found is that for my printer, the tension between the left and right belts is slightly different (counter to conventional wisdom).

15

Re: Done all calibrations, still no perfect circles

Shotline: I meant the rods may be bent from the factory as there have been quite a few users that have noticed this on the z axis, I think it is from how someone is cutting the rods or perhaps they are just getting bent in shipment prior to assembly??? I personally know this can happen as I order drillrod every so often and I have quite a bit of rods arrive bent... some hardly noticeable and a few that look like a skinny banana...

One other thing to be careful of is tension, if you tension the belts too tight it will stretch the belt and in turn give you more backlash as the teeth will have more space between them and will not mesh correctly with the pulleys.